CASE NOTE
ZENTAII AND THE TROUBLES OF EXTRADITION
MATTHEW STUBBS*
In Minister for Home Affairs (Cth) v Zentai, the High Court held that
the Treaty on Extradition between Australia and the Republic of Hungary
prevented Mr Zentai’s extradition to face accusations of committing a
retrospective war crimes offence. The case is, in all likelihood, the last
episode in the history of Australia’s contributions to bringing to justice
alleged war criminals from the Second World War. The extraordinary time
and resources devoted to the ultimately abortive extradition process in
Zentai raise questions about the efficiency of the procedures under the
Extradition Act 1988 (Cth). Further, given the strict textual interpretation
adopted by the Court, amendment to some of Australia’s extradition treaties
may be required to uphold the key purpose of extradition arrangements,
which is to facilitate international cooperation in the apprehension and
surrender for trial of those accused of serious criminal offences.
I
INTRODUCTION
The High Court in Minister for Home Affairs (Cth) v Zentai1 was required by
the provisions of the Extradition Act 1988 (Cth) to interpret and apply the 1997
Treaty on Extradition between Australia and the Republic of Hungary.2 Part II
of this case note begins by examining the law applicable to Hungary’s request
for the extradition of Mr Zentai. Part III catalogues the extraordinary extradition
process in Zentai and reflects on its significance for potential reform of extradition
arrangements. Part IV then analyses the Court’s interpretation of art 2.5(a) of
the Australia-Hungary Extradition Treaty which led to the result that Mr Zentai
cannot be extradited to Hungary to face investigation for alleged war crimes
committed during the Second World War. As will be seen, the Court’s strict
textual approach to the relevant provisions of the Australia-Hungary Extradition
Treaty may have significant implications for future extradition cases, and may
require remedial amendments to some of Australia’s extradition treaties.
*
BFin, BEc, LLB(Hons), GDLP, PhD; Barrister and Solicitor of the Supreme Court of South Australia
and the High Court of Australia; Senior Lecturer, Adelaide Law School, University of Adelaide.
1
2
(2012) 246 CLR 213 (‘Zentai’).
Treaty on Extradition between Australia and the Republic of Hungary, signed 25 October 1995, [1997]
ATS 13 (entered into force 25 April 1997) (‘Australia-Hungary Extradition Treaty’), reproduced in
Extradition (Republic of Hungary) Regulations 1997
7 (Cth) sch 1.
895
Zentai and The Troubles of Extradition
II
THE APPLICABLE LAW OF EXTRADITION
The Australia-Hungary Extradition Treaty was given effect by the Extradition
Act 1988 (Cth) (‘Extradition
‘
Act’) and the Extradition (Republic of Hungary)
Regulations 1997
7 (Cth).3 The Extradition Actt prescribes a four-step process for
extradition.4 First, the country requesting extradition may seek an extradition
arrest warrant from a Magistrate under s 12. Second, s 16 requires the AttorneyGeneral to consider whether to issue a notice that will bring on an extradition
hearing (a decision not to issue a notice will terminate the extradition process).5
Third, s 19 provides for an extradition hearing before a Magistrate who will
‘determine whether the person is eligible for surrender’.6 Fourth, s 22 gives the
Attorney-General the final choice to ‘determine whether the person is to be
surrendered’.7 The Attorney-General had delegated these powers to the Minister
for Home Affairs.8
The High Court’s decision in Zentai concerned the fourth step. Under s 22(2) of the
Extradition Act, the Minister’s power to determine that a person be surrendered
was subject to the conditions listed in s 22(3). Relevantly, s 22(3)(e) treated art
2.5(a) of the Australia-Hungary Extradition Treaty as ‘a limitation, condition,
qualification or exception’ restricting extradition, and therefore required that the
Minister be satisfied that the circumstance did not exist before he could determine
that Mr Zentai be surrendered to Hungary. Consequently, the Minister could only
determine to extradite Mr Zentai under s 22 of the Extradition Act, which picked
up art 2.5(a) of the Australia-Hungary Extradition Treaty, if the offence alleged
against Mr Zentai was ‘an offence in the Requesting State at the time of the acts
or omissions constituting the offence’.
III
THE EXTRADITION OF MR ZENTAI
A
Procedural History
On 23 March 2005, Hungary requested that Mr Zentai be extradited to face
questioning in relation to an alleged war crime (the murder of a young Jewish
man) in 1944.9 The war crimes offence had been created in 1945, although
murder was a crime in Hungary in 1944.10 It is not clear why Hungary requested
extradition on the basis of the retrospective war crimes offence rather than the
3
4
5
6
7
8
9
10
These regulations were made under the Extradition Actt s 11(1)(a).
For an explanation of the procedure, see E P Aughterson, Extradition: Australian Law and Procedure
(Law Book, 1995) 30–1, 183–235.
Extradition Actt s 17.
Ibid s 19(1) (emphasis added).
Ibid s 22(2) (emphasis added).
Zentai (2012) 246 CLR 213, 233 [50] (Gummow, Crennan, Kiefel and Bell JJ).
Ibid 219 [7]–[8] (French CJ).
Ibid 219–20 [7]–[9].
896
Monash University Law Review (Vol 39, No 3)
offence of murder, which would have been applicable prospectively. On 8 July
2005, the Minister gave notice under s 16 of the Extradition Actt in relation to Mr
Zentai,11 triggering an extradition hearing before a state Magistrate under s 19.
Mr Zentai challenged the Magistrate’s power to conduct this hearing, but failed
before the Federal Court,12 and his appeals to the Full Court of the Federal Court13
and, by special leave,14 the High Court,15 were dismissed. On 20 August 2008,
a Magistrate determined that Mr Zentai was eligible for extradition under s 19
of the Extradition Act.16 Mr Zentai then sought a review of the Magistrate’s
determination before the Federal Court, but the Magistrate’s determination
was upheld17 and Mr Zentai’s appeal to the Full Court of the Federal Court was
dismissed.18
On 12 November 2009, the Minister determined under s 22 of the Extradition
Actt that Mr Zentai should be extradited to Hungary.19 Mr Zentai challenged this
determination before the Federal Court, and was successful.20 The Minister’s
appeal to the Full Court of the Federal Court was only successful in part: the Full
Court did not displace the ruling against extradition made by McKerracher J.21
The Minister’s further appeal,22 by special leave to the High Court in Zentai,23
was dismissed.24
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Mr Zentai was arrested and then bailed: see Zentai v O’Connor [No 3] (2010) 187 FCR 495, 502
[9]–[11].
Zentai v Hungary (2006) 153 FCR 104. Mr Zentai then obtained orders restraining the Republic of
Hungary from further proceeding with his extradition: Zentai v Hungary [No 2] [2006] FCA 1735 (30
November 2006). While awaiting a special leave hearing in the High Court, Mr Zentai unsuccessfully
sought a further order restraining the Republic of Hungary from prosecuting eligibility proceedings:
Zentai v Hungary [2007] FCA 842 (29 May 2007).
Zentai v Hungary (2007) 157 FCR 585.
See Transcript of Proceedings, Zentai v Hungary [2007] HCATrans 491 (3 September 2007).
O’Donoghue v Ireland
d (2008) 234 CLR 599.
Zentai (2012) 246 CLR 213, 220 [10] (French CJ). Mr Zentai obtained bail from the Federal Court that
same day: Zentai v Hungary [2008] FCA 1335 (20 August 2008).
Zentai v Hungary [2009] FCA 284 (31 March 2009). Mr Zentai then obtained further bail from the
Federal Court: Zentai v Hungary [2009] FCA 511 (12 May 2009).
Zentai v Hungary (2009) 180 FCR 225.
Zentai (2012) 246 CLR 213, 217–18 [2] (French CJ). Mr Zentai then again obtained bail in the Federal
Court: Zentai v O’Connorr (2009) 263 ALR 511.
Zentai v O’Connor [No 3] (2010) 187 FCR 495; Zentai v O’Connor [No 4] [2010] FCA 1385 (10
December 2010). See Stephen Tully, ‘Zentai v Honourable Brendan O’Connor [No 3] [2010] FCA 691
(2 July 2010)’ (2010) 17 Australian International Law Journall 267. In the interim, Mr Zentai had taken
action to resist a claim of legal professional privilege over certain documents: Zentai v O’Connor [No 2]
(2010) 183 FCR 180.
O’Connor v Zentai (2011) 195 FCR 515.
After a further hearing in the Federal Court regarding stays: O’Connor v Zentai [2011] FCA 1162
(5 October 2011).
Transcript of Proceedings, Minister for Home Affairs (Cth) v Zentai [2011] HCATrans 339 (9 December
2011).
Further detail about the procedural history of the case, and exchanges between representatives of Mr
Zentai and the Commonwealth, is given in: Zentai v O’Connor [No 3] (2010) 187 FCR 495, 501–6
[4]–[65].
Zentai and The Troubles of Extradition
897
B Reflections on the Extradition Process in Light of Zentai
Mr Zentai’s extradition commenced with a warrant issued in Hungary on 3 March
2005,25 and concluded with the High Court’s decision in Zentai on 15 August
2012, after more than seven years and five months, during which time nineteen
different Australian judicial decisions were required, in addition to the Minister’s
notice under s 16 and determination under s 22 of the Extradition Act. All of this
notwithstanding that ‘extradition proceedings do not involve determination of the
question of guilt’.26 Indeed, consistent with the predominant ‘no evidence’ model
of extradition in Australia, there was not even a judicial assessment of whether
there was a prima facie case against Mr Zentai.27
It should be noted that the 2012 amendments to the Extradition Actt28 do not
directly remedy any of the difficulties encountered in Zentai. The amendments
do clarify provisions regarding bail,29 which was an issue addressed on a number
of occasions regarding Mr Zentai,30 but was not a matter that delayed the course
of his extradition process.
The extradition process might be improved by bringing within the scope of the
Magistrate’s s 19 determination the restraints preventing extradition that are
currently matters for the Minister’s satisfaction under s 22(3)(e). However, the
choice to leave this matter to executive discretion rather than judicial decision was
deliberate. In 2001, the Joint Standing Committee on Treaties received evidence
from commentators in support of transferring these issues to the courts,31 but
also received evidence in favour of the status quo.32 The Committee reported that
‘concerns expressed about the way in which the Act has placed responsibility for
scrutiny of human rights protections in the hands of the executive rather than the
courts have some force’, but in the end merely recommended the issue ‘be further
explored’.33
In Zentai, the critical issue regarding art 2.5(a) of the Australia-Hungary
Extradition Treaty would have arisen at an earlier stage in the process if such
protections were a matter for the s 19 determination of the Magistrate, rather
than for the Minister’s satisfaction. However, incorporation of such issues at the
25
26
27
28
29
30
31
32
33
Zentai (2012) 246 CLR 213, 219 [7] (French CJ).
I A Shearer, Extradition in International Law (Manchester University Press, 1971) 146.
As to the ‘no evidence’ model of extradition, see, eg, Aughterson, above n 4, 217–23; Gavan Griffith
and Claire Harris, ‘Recent Developments in the Law of Extradition’ (2005) 6 Melbourne Journal of
International Law 33, 53; Vasiljkovic v Commonwealth (2006) 227 CLR 614, 619–22 [9]–[15], 626
[23], 631–3 [39]–[41] (Gleeson CJ), 633 [44], 643 [87], 644–7 [91]–[105] (Gummow and Hayne JJ),
with whom Heydon J agreed: at 676 [222] (‘Vasiljkovic’). Cf 651–2 [123]–[124], 658–65 [146]–[169],
667 [179], 672–3 [204]–[210] (Kirby J). See also Sascha Morrell, ‘Vasiljkovic v Commonwealth of
Australia’ (2007) 29 Sydney Law Review 321, 333–6.
Extradition and Mutual Assistance in Criminal Matters Legislation Amendment Act 2012 (Cth).
Ibid sch 2 pt 3 div 8.
See above nn 16–17, 19.
Joint Standing Committee on Treaties, Parliament of Australia, Report 40: Extradition — A Review of
Australia’s Law and Policy (2001) 55–6 [4.7]–[4.11].
Ibid 56–7 [4.12]–[4.16].
Ibid 57 [4.17].
898
Monash University Law Review (Vol 39, No 3)
determination stage would inevitably lengthen the time taken to reach that point
in the extradition process. It might also call for evidence to be led, which would
be inconsistent with the overall ‘no evidence’ model of extradition.
Zentai is an exceptional case34 where extraordinary resources were devoted to
a legal process which in the end concluded without any determination of Mr
Zentai’s guilt or innocence. Although Australia’s current system of extradition
has been criticised on the basis that it ‘subordinates individual rights to
administrative efficiency’,35 the procedural history of Zentai reminds us again
that the reverse is sometimes true.36 The ‘no evidence’ model of extradition has
already been adopted with a view to streamlining the extradition process. It is a
difficult question whether further steps could be taken to promote efficiency in
the extradition process without compromising individual rights, but nonetheless
one worthy of further investigation.
IV THE ZENTAII REASONING: INTERPRETING ART 2.5(a)
As explained above, the critical issue in Zentai was whether the war crimes
offence was ‘an offence in the Requesting State at the time of the acts or omissions
constituting the offence’ given it had been created in 1945, albeit that murder
had been a crime in Hungary in 1944.37 Article 2 of the Australia-Hungary
Extradition Treaty, headed ‘Extraditable Offences’, contained the following
relevant provisions:
1.
For the purposes of this Treaty, extraditable offences are offences
however described which are punishable under the laws of both
Contracting States by imprisonment for a maximum period of at
least one year or by a more severe penalty. …
2.
For the purpose of this Article in determining whether an offence is
an offence against the law of both Contracting States:
(a)
34
35
36
37
it shall not matter whether the laws of the Contracting States
place the acts or omissions constituting the offence within the
same category of offence or denominate the offence by the
same terminology;
One commentary devotes an entire chapter to extradition of alleged war criminals, because ‘[w]ar
criminals are sui generis’: Geoff Gilbert, Aspects of Extradition Law (Martinus Nijhoff, 1991) ch 8.
Morrell, above n 27, 338, citing Human Rights and Equal Opportunity Commission, Submission to the
Attorney-General’s Department on the Extradition Review Discussion Paperr (2006) [14.5].
As Gilbert has observed,
[t]he delicate balance seen in extradition law between, on the one hand, its provision of mutual
assistance in criminal matters at an international level between states and, on the other, its
protection of the rights of the fugitive, means that occasionally there will be a conflict and that
one party will suffer.
Gilbert, above n 34, 245.
Australia-Hungary Extradition Treaty art 2.5(a).
Zentai and The Troubles of Extradition
(b)
5.
899
the totality of the acts or omissions alleged against the person
whose extradition is sought shall be taken into account and it
shall not matter whether, under the laws of the Contracting
States, the constituent elements of the offence differ. …
Extradition may be granted pursuant to the provisions of this Treaty
irrespective of when the offence in relation to which extradition is
sought was committed, provided that:
(a)
it was an offence in the Requesting State at the time of the acts
or omissions constituting the offence; and
(b)
the acts or omissions alleged would, if they had taken place in
the territory of the Requested State at the time of the making of
the request for extradition, have constituted an offence against
the law in force in that State.
Article 2.1 states a rule of double criminality,38 which art 2.2 ensures is applied
flexibly, taking into account not merely the offence but ‘the totality of the acts or
omissions alleged’.39 The purpose of art 2.5(a) is less obvious. The critical issue
in Zentai was whether this provision permitted extradition if the alleged conduct
constituted an offence, meaning any offence known to law, at the time of it being
committed (the broad view); or alternatively, whether extradition was permitted
only if the specific offence, with which the accused is charged, was in existence
at the time of the alleged conduct (the narrow view).
The Minister argued for the broader interpretation, under which art 2.5(a) would
not prevent extradition, because Mr Zentai’s alleged conduct constituted murder
and was therefore an offence at the time of its occurrence in 1944. However, Mr
Zentai advocated the narrower view: given that the war crime alleged against him
did not come into existence until 1945, it was not ‘an offence in the Requesting
State at the time’ of the acts in 1944, and could not satisfy the requirement
expressed in art 2.5(a) of the Australia-Hungary Extradition Treaty.
The analysis of the Court’s reasoning in Zentai is organised as follows. First,
it investigates the Court’s approach to the object and purpose of the AustraliaHungary Extradition Treaty as a whole, and more specifically to the object and
purpose of art 2.5(a) in the context of international law’s prohibition of retrospective
criminal laws. Second, it examines the predominantly textual interpretation of
the Justices. Finally, consideration is given to two subsidiary issues: the relevance
of the subsequent agreement and practice of the treaty parties to the interpretation
of the extradition treaty, and the potential applicability of a presumption in favour
of individual liberty.
38
39
See, eg, Gillian D Triggs, International Law: Contemporary Principles and Practices (LexisNexis
Butterworths, 2ndd ed, 2011) 465 [8.46]; Shearer, Extradition in International Law, above n 26, 137–41;
Aughterson, above n 4, 59–69; M Cherif Bassiouni, International Extradition and World Public Order
(A W Sijthoff-Leyden, 1974) 322–6; Griffith and Harris, above n 27, 37–8; Gilbert, above n 34, 47–8.
See, eg, Shearer, Extradition in International Law, above n 26, 141–7; Aughterson, above n 4, 69–71,
78–80; Gilbert, above n 34, 48–52.
900
Monash University Law Review (Vol 39, No 3)
A The Purpose of the Extradition Treaty and Art 2.5(a)
Australia is a largely dualist nation, following the common law tradition in
treating international law as a separate system whose rules are not automatically
enforceable in Australia.40 However, where international law has been incorporated
by legislation, the High Court has endorsed the use of the international law of
treaty interpretation in the interpretation of such legislation.41 This requires an
interpretation which takes into account the ‘object and purpose’ of the treaty.42
In Zentai, this approach was expressly endorsed by French CJ.43 The purpose
of extradition arrangements is reflected in his Honour’s statement that the
extradition treaty ‘reflects, according to its recital, the desire of both countries to
make more effective their cooperation in the suppression of crime’.44 However, as
will be seen below, French CJ did not expressly apply this object and purpose to
the interpretation of the Australia-Hungary Extradition Treaty.
The joint judgment of Gummow, Crennan, Kiefel and Bell JJ merely stated:
‘Consideration of the object and purpose of the Treaty does not assist in
ascertaining the meaning of [art 2.5(a)]’.45 Their Honours disapproved of North
J’s dissent in the Full Court of the Federal Court, which relied on the AustraliaHungary Extradition Treaty’s purpose as being ‘to ensure that people are called
to account for their wrongdoing’.46 Instead, the joint judgment referred to the art
1 undertaking ‘to extradite … any person … who is wanted for prosecution by
a competent authority for … an extraditable offence against the law of the other
40
41
42
43
44
45
46
The dualist approach to incorporation is described by Lord Atkin in A-G (Canada) v A-G (Ontario)
[1937] AC 326, 347: ‘the making of a treaty is an executive act, while the performance of its obligations,
if they entail alteration of the existing domestic law, requires legislative action’. See, eg, Hilary
Charlesworth et al, No Country Is an Island: Australia and International Law (UNSW Press, 2006)
29; I A Shearer, ‘The Relationship between International Law and Domestic Law’ in Brian R Opeskin
and Donald R Rothwell (eds), International Law and Australian Federalism (Melbourne University
Press, 1997) 34, 52–3; Triggs, above n 38, 177–9 [4.45]–[4.46]; Donald K Anton, Penelope Mathew
and Wayne Morgan, International Law: Cases and Materials (Oxford University Press, 2005) 406, 426;
Alexander Reilly et al, Australian Public Law (Oxford University Press, 2011) 216–17, 236–7.
See, eg, Povey v Qantas Airways Ltd
d (2005) 223 CLR 189, 202 [24] (Gleeson CJ, Gummow, Hayne and
Heydon JJ), 211 [60] (McHugh J), 232 [136], 234 [142] (Kirby J); D C Pearce and R S Geddes, Statutory
Interpretation in Australia (LexisNexis Butterworths, 7th ed, 2011) 43–6 [2.20]–[2.21]; Stephen Hall,
Principles of International Law (LexisNexis Butterworths, 3rdd ed, 2011) 175–6 [3.39].
Vienna Convention on the Law of Treaties, opened for signature 23 May 1969, 1155 UNTS 331 (entered
into force 27 January 1980) art 31(1) (‘VCLT’): ‘A treaty shall be interpreted in good faith in accordance
with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its
object and purpose’.
Zentai (2012) 246 CLR 213, 222–3 [17]–[19]. His Honour referred inter alia to: Shipping Corporation
of India Ltd v Gamlen Chemical Co (A/asia) Pty Ltd
d (1980) 147 CLR 142, 159 (Mason and Wilson JJ),
quoting James Buchanan & Co Ltd v Babco Forwarding & Shipping (UK) Ltd
d [1978] AC 141, 152
(Lord Wilberforce); Applicant A v Minister for Immigration and Ethnic Affairs (1997) 190 CLR 225,
240 (Dawson J), 254 (McHugh J), quoted in Zentai (2012) 246 CLR 213, 223 [19]. As G R Kennett
SC (for Mr Zentai) had accepted, ‘[o]ne gives the treaty, in its domestic application, the meaning that it
would be given if it needed to be construed as a matter of international law’: Transcript of Proceedings,
Minister for Home Affairs v Zentai [2012] HCATrans 82 (28 March 2012) (G R Kennett SC).
Zentai (2012) 246 CLR 213, 224 [20].
Ibid 239 [65].
O’Connor v Zentai (2011) 195 FCR 515, 521 [27], quoted in Zentai (2012) 246 CLR 213, 237 [63]
(Gummow, Crennan, Kiefel and Bell JJ). See also at 238 [65].
Zentai and The Troubles of Extradition
901
Contracting State’. On this basis, their Honours indicated that the purpose of
the extradition treaty ‘is to give effect to the reciprocal obligations to extradite
persons for extraditable offences’.47 It is unclear how this very narrow statement
of purpose advances understanding of the text.
The majority’s narrow perspective on the purpose of the Australia-Hungary
Extradition Treaty may be contrasted with the broader understanding of the
purpose of extradition displayed in the judgment of Gleeson CJ in Vasiljkovic,
where his Honour wrote that:
[Extradition] is undertaken for the purpose of enabling … an adjudication
[of guilt or innocence] to be made in a foreign place, according to foreign
law, in circumstances where Australia has no intention itself of bringing
the person to trial for the conduct of which the person is accused.48
Indeed, the positive role of extradition in the promotion of international justice
has long been accepted by the High Court. As Barwick CJ observed in Barton
v Commonwealth, ‘[t]he co-operation of nations in the surrender of fugitives
from justice is a most important aspect of international life’.49 The importance
of extradition was further noted in United Mexican States v Cabal, where
Gleeson CJ, McHugh and Gummow JJ warned that: ‘In an era where much crime
is transnational, the breakdown of international co-operation in apprehending
criminals would be disastrous for the peoples of the countries concerned’.50 This
is not to say that the protection of the rights of individuals in the extradition
process is unimportant, but it is a reminder that, as Gummow and Hayne JJ
noted in Vasiljkovic, quoting from Quick and Garran, extradition treaties and
their implementing legislation serve the ‘common interest’ of ‘all civilized
communities … in the administration of the criminal law and in the punishment
of wrongdoers’51 — an object which is obscured by a mere statement that the
purpose of the extradition treaty is to extradite persons for extraditable offences.
When interpreting the Australia-Hungary Extradition Treaty in Zentai, their
Honours neglected extradition’s role in promoting international cooperation in
the enforcement of criminal justice. To the extent that an object and purpose was
referred to, it was in the context of art 2.5(a) alone.52
The majority identified the purpose of art 2.5(a) as the prevention of retrospective
criminal liability. The joint judgment of Gummow, Crennan, Kiefel and Bell JJ
stated: ‘Article 2.5(a) reflects the adoption by Australia and Hungary of a policy
47
48
49
50
51
52
Zentai (2012) 246 CLR 213, 238–9 [65] (emphasis in original).
(2006) 227 CLR 614, 629–30 [34].
(1974) 131 CLR 477, 483.
(2001) 209 CLR 165, 190 [58].
Vasiljkovic (2006) 227 CLR 614, 633 [45], quoting John Quick and Robert Randolph Garran, The
Annotated Constitution of the Australian Commonwealth (Angus and Robertson, 1901) 635.
As Young CJ in Eq cautioned in respect of corporations legislation in Edwards v A-G (NSW) (2004) 60
NSWLR 667, 681: ‘whenever one is construing any particular provision one must be careful not to take
one’s gaze off the essential purpose [of the Act as a whole] and pay overmuch attention to technical
details of wording of individual provisions’. See, eg, Pearce and Geddes, above n 41, 36–40 [2.11]–
[2.13].
902
Monash University Law Review (Vol 39, No 3)
against the imposition of criminal liability or punishment retrospectively’.53 For
French CJ, similarly, art 2.5(a) ‘gives practical effect to the general principle
against retroactive municipal criminal law’.54 Their Honours referred to no
authority to establish that this was the purpose of the provision, which may reflect
a conclusion that the text alone gave a clear indication of art 2.5(a)’s purpose,
notwithstanding that this issue in fact divided the Justices of the Federal Court
and High Court.
The following exchanges that occurred at the hearing of Zentai (but which are not
referred to in the judgments) may provide an insight into their Honours’ thinking:
GUMMOW J: … I suspect Australia now has, and this is one of them,
extradition treaties with a number of former totalitarian countries in
Europe […], of which Hungary would be one, made after the collapse of
Communism. Is there any evident practice of inserting in those extradition
treaties with countries which, under their totalitarian systems, might be
expected to have had retrospective criminality, something like 5(a)?
…
MR LLOYD: … I have provided to the Court … a reference to countries
that have similar treaties in similar terms. … [I]nsofar as your Honour
Justice Gummow asked for a list of totalitarian countries with which
we have agreements, this is apropos a more general response as to the
countries with which we have agreements.
HEYDON J: Apart from Switzerland they all seem totalitarian.55
These exchanges raise the possibility that their Honours concluded that art 2.5(a)
was inserted to restrict extradition in the case of formerly ‘totalitarian’ countries
which might have objectionable retrospective criminal laws, although such a
conclusion is speculative.
The information before the Court regarding the purpose of art 2.5(a) was limited.
Reference was made to the (first ever) report of the Joint Standing Committee
on Treaties,56 which briefly commented on the ‘current group of extradition
agreements’57 (including the Australia-Hungary Extradition Treaty),
y identifying
as ‘protections available to potential extraditees’58 both the requirement of double
criminality and what the report refers to as the ‘timing test’ which ‘requires
that the conduct which is subject of extradition must constitute an offence in the
53
54
55
56
57
58
Zentai (2012) 246 CLR 213, 241 [70].
Ibid 228 [32]. At first instance in the Federal Court, McKerracher J similarly found that art 2.5(a) was
‘directed to excluding … cases of foreign legislation with retrospective application’: Zentai v O’Connor
[No 3] (2010) 187 FCR 495, 541 [190].
Transcript of Proceedings, Minister for Home Affairs (Cth) v Zentai [2012] HCATrans 82 (28 March
2012).
On the 1996 treaty-making reforms, see generally Anton, Mathew and Morgan, above n 40, 428–
43; Donald R Rothwell et al, International Law: Cases and Materials with Australian Perspectives
(Cambridge University Press, 2011) 186–9; Charlesworth et al, above n 40, 41–8.
Joint Standing Committee on Treaties, Parliament of Australia, First Reportt (1996) 9 [2.5].
Ibid.
Zentai and The Troubles of Extradition
903
Requesting State at the time it occurred’.59 This is merely a restatement of the
words of art 2.5(a). At its highest, it provides limited support for the provision’s
purpose as protecting
g potential extraditees.
The lone dissentient, Heydon J, did not find that the purpose of art 2.5(a) was
to protect individuals from extradition to face retrospective criminal offences.
Indeed, his Honour was skeptical about international law’s resistance to
retrospective criminality more generally:
Counsel for the first respondent referred to the ‘general antipathy in
international law … against retrospectivity’ — an antipathy which can be
somewhat selectively displayed. Analysis should not be diverted by that
antipathy in this case. Nor … by reflections upon the zeal with which the
victors at the end of the Second World War punished the defeated for war
crimes. The victors were animated by the ideals of the Atlantic Charter
and of the United Nations. The Universal Declaration of Human Rights
was about to peep over the eastern horizon. But first, they wanted to have
a little hanging.60
On his Honour’s last point, it might be noted that objections to the sentences
handed down by Australian war crimes trials of Japanese defendants after the
second world war often related to the leniency, not harshness, of sentencing.61
There was comparatively little hanging,62 and considerable mercy was shown to
those imprisoned.63
The Court’s treatment of international law’s prohibition of retrospective
criminal laws was brief, this purposive approach being subordinated to a textual
interpretation of the Australia-Hungary Extradition Treaty. However, it is worth
further exploring this prohibition, which is stated in art 15 of the International
Covenant on Civil and Political Rights.64 Even setting aside the exception for
international crimes in art 15(2) which was not argued in Zentai (but might have
59
60
61
62
63
64
Ibid 9 [2.6].
Zentai (2012) 246 CLR 213, 243 [75].
See, eg, Emmi Okada, ‘The Australian Trials of Class B and C Japanese War Crime Suspects, 1945–51’
(2009) 16 Australian International Law Journall 47, 79. Okada concludes that ‘[w]hile the Australian
military tribunal meted out some harsh sentences, other cases showed a surprising degree of leniency’.
Of 924 Japanese tried by Australian military courts, 280 (30 per cent) were acquitted. Of the 644
convicted, 148 (23 per cent) were sentenced to death, the rest to imprisonment: see Gavin Long, ‘The
Final Campaigns’ in Australia in the War of 1939–1945: Series 1 — Army (Australian War Memorial,
1963) vol VII, 583 n 9; D C S Sissons, ‘The Australian War Crimes Trials and Investigations (1942–51)’
19–20 <http://www.ocf.berkeley.edu/~changmin/documents/Sissons%20Final%20War%20Crimes%20
Text%2018-3-06.pdf>.
Many of the sentences of imprisonment (221 of 496, or 45 per cent) were for less than 10 years, and in
fact only 14 prisoners remained in custody on 4 July 1957 when these final prisoners were released: see
Long, above n 62, 583 n 9; Sissons, above n 62, 19–20, 53.
International Covenant on Civil and Political Rights, opened for signature 16 December 1966, 999
UNTS 171 (entered into force 23 March 1976) (‘ICCPR’).
904
Monash University Law Review (Vol 39, No 3)
been relevant,65 given the locus classicus of the principle it represents is the war
crimes judgments at Nuremberg),66 there is a good argument that art 15(1) was not
breached in Mr Zentai’s situation.
Article 15(1) of the ICCPR states:
No one shall be held guilty of any criminal offence on account of any act or
omission which did not constitute a criminal offence … at the time when
it was committed. Nor shall a heavier penalty be imposed than the one
that was applicable at the time when the criminal offence was committed.
This prohibition of retrospective criminal laws implements two fundamental
maxims.67 First, nullum crimen sine lege, meaning that conduct which was legal
when it occurred should not be retrospectively criminalised.68 Second, nulla
poena sine lege, meaning that liability to punishment should not be retrospectively
increased.69 Together, these maxims ensure that no individual is convicted of
a criminal offence for conduct which was legal when committed, and that no
individual receives a harsher punishment when convicted than applied at the time
the offence was committed. Therefore, two critical questions would need to be
65
66
67
68
69
See Human Rights Committee, Views: Communication No 960/2000, 78th sess, UN Doc
CCPR/C/78/D/960/2000 (19 September 2003) [9.3] (‘Baumgarten v Germany’); Sarah Joseph and
Melissa Castan, The International Covenant on Civil and Political Rights: Cases, Materials and
Commentary (Oxford University Press, 3rdd ed, 2013) 529–32 [15.16]–[15.18]; Hall, above n 41, 563–4
[11.61].
Adam McBeth, Justine Nolan and Simon Rice, The International Law of Human Rights (Oxford
University Press, 2011) 95–6.
‘Basic concepts of fairness and justice require that one cannot be punished for something which was
not a crime at the time it was committed — nullum crimen sine lege and nulla poena sine lege’: Rhona
K M Smith, Textbook on International Human Rights (Oxford University Press, 4th ed, 2010) 260. See
also Joseph and Castan, above n 65, 521 [15.01]; Manfred Nowak, UN Covenant on Civil and Political
Rights: CCPR Commentary (N P Engel, 2ndd revised ed, 2005) 359; Sangeeta Shah, ‘Administration of
Justice’ in Daniel Moeckli, Sangeeta Shah and Sandesh Sivakumaran (eds), International Human Rights
Law (Oxford University Press, 2010) 304, 328. On the evolution of these maxims and their application
in international criminal law, see generally M Cherif Bassiouni, ‘Principles of Legality in International
and Comparative Criminal Law’ in M Cherif Bassiouni (ed), International Criminal Law (Martinus
Nijhoff, 3rdd ed, 2008) vol 1, 73. These maxims also form the foundation of the equivalent prohibition
of retrospective criminal law in the Convention for the Protection of Human Rights and Fundamental
Freedoms, opened for signature 4 November 1950, 213 UNTS 221 (entered into force 3 September
1953) art 7: see P van Dijk and G J H van Hoof, Theory and Practice of the European Convention on
Human Rights (Kluwer, 2ndd ed, 1990) 359; D J Harris, M O’Boyle and C Warbrick, Law of the European
Convention on Human Rights (Butterworths, 1995) 274; J E S Fawcett, The Application of the European
Convention on Human Rights (Clarendon Press, 1969) 178; Francis G Jacobs and Robin C A White, The
European Convention on Human Rights (Clarendon Press, 2ndd ed, 1996) 162.
This is described as ‘a guarantee to individuals that they will not be held accountable … for actions
which were not deemed criminal at the time of commission or omission’ in Smith, above n 67, 261.
For the comment that it ensures accountabtility for ‘no crime except in accordance with the law’, see
Joseph and Castan, above n 65, 521 [15.01]. Similarly, Bassiouni says that ‘a crime must be sufficiently
defined to put people on notice that a particular conduct has been characterized as criminal’: Bassiouni,
‘Principles of Legality’, above n 67, 100. See also Nowak, above n 67, 359–62; Shah, above n 67, 328.
This is described as ‘no punishment except in accordance with the law’: Joseph and Castan, above
n 65, 521 [15.01]. Similarly, Nowak says that ‘a penalty heavier than the one that was applicable at
the time the offence was committed may not be imposed’: Nowak, above n 67, 363. For consideration
of this principle, see M J Bossuyt, Guide to the ‘Travaux Préparatoires’ of the International Covenant
on Civil and Political Rights (Martinus Nijhoff, 1987) 323; Joseph and Castan, above n 65, 524–6
[15.07]–[15.12]; Nowak, above n 67, 363–5; Shah, above n 67, 328.
Zentai and The Troubles of Extradition
905
answered to apply the prohibition of retrospective criminal laws to Mr Zentai’s
case. First, was Mr Zentai’s conduct legal when committed, but then retrospectively
criminalised? Second, if not, was any applicable punishment retrospectively
increased? If neither of these applies, the prohibition of retrospective criminal
law would not apply to Mr Zentai.70
In similar circumstances, the Human Rights Committee held art 15(1) of the
ICCPR inapplicable in Westerman v Netherlands.71 The author had been charged
with a new retrospective offence that did not precisely correspond to a previous
offence in force at the time of the conduct. The act when committed by Westerman
was criminal (as ‘refuses or intentionally fails to obey any official order’), and the
act as charged was criminal (as ‘refuses or intentionally fails to perform any
duty’), but they were clearly different offences: ‘the nature of the offence … was
different’.72 The Committee found that there was no breach of art 15(1), despite
the later offence being different, because the relevant ‘acts were an offence at the
time they were committed’.73
The resemblance of these facts to Zentai is striking: they suggest that where acts
are criminal when committed and criminal when convicted, even if the actual
offences and their elements differ, there is no violation of art 15(1), so long as
there is no increased punishment.74 Such was the situation of Mr Zentai: wanted
now for a retrospective war crimes offence in respect of conduct that could have
been charged as a murder under prospective criminal law. Moreover, both the war
crimes and murder offences attracted a maximum penalty of death at the time in
Hungary, which had subsequently been converted to a maximum penalty of life
70
71
72
73
74
In the Federal Court, McKerracher J had used the two maxims to interpret art 2.5(a) of the AustraliaHungary Extradition Treaty: Zentai v O’Connor [No 3] (2010) 187 FCR 495, 541–2 [191]. However,
his Honour referred only to arts 22 and 23 of the Rome Statute of the International Criminal Court,
opened for signature 17 July 1998, 2187 UNTS 90 (entered into force 1 July 2002) (‘Rome Statute’),
which state the maxims in defining the jurisdiction of the International Criminal Court, before applying
the principles more strictly than they are applied under international law, writing that:
This presupposes that the offence of war crime was both clearly defined in the relevant
Hungarian written law and that the penalty was publicised in that statute or decree. Both
those conditions were absent on 8 November 1944 when the ‘crime’ was alleged to have been
committed.
Zentai v O’Connor [No 3] (2010) 187 FCR 495, 541–2 [191]–[192]. His Honour’s approach treats
the prohibition of retrospective criminal law as absolute, rather than as directed to ensuring that
legal conduct is not retrospectively criminalised and that offences are not retrospectively subjected
to increased punishment. A similar approach was taken by Jessup J in the Full Court of the Federal
Court, again without reference to any authority on the international human rights prohibition of
retrospective criminal laws: O’Connor v Zentai (2011) 195 FCR 515, 571 [157]–[158]. It may be that
different considerations in respect of the prohibition of retrospective criminal laws arise in the context
of extradition, although there is no present authority addressing this point.
Human Rights Committee, Views: Communication No 682/1996, 67th sess, UN Doc
CCPR/C/67/D/682/1996 (13 December 1999) (‘Westerman v Netherlands’).
Nowak, above n 67, 364.
Westerman v Netherlands, UN Doc CCPR/C/67/D/682/1996, 6 [9.2] (emphasis added). See also Joseph
and Castan, above n 65, 524 [15.06].
See, eg, Nowak, above n 67, 364–5.
906
Monash University Law Review (Vol 39, No 3)
imprisonment,75 so no greater punishment arose under the war crimes charge
than under a charge of murder. To answer the two questions identified above:
the conduct was illegal (as murder) when committed, and attracted a penalty of
equal severity for either offence. Accordingly, were Mr Zentai to be convicted of
the war crimes offence for which his extradition was sought, there would be no
breach of art 15(1) of the ICCPR.
The Court’s use of the object and purpose of the Australia-Hungary Extradition
Treaty was limited. The Justices did not draw upon extradition’s overall purpose
of promoting international cooperation in the enforcement of criminal justice.
However, the Justices took into consideration the avoidance of retrospective
criminal liability in interpreting art 2.5(a), notwithstanding a lack of evidence
about the purpose of this provision (beyond the text itself). Moreover, their
Honours did not examine whether the international prohibition of retrospective
criminal law would be violated; in fact, it is likely that it would not be. Overall,
the judgments in Zentai made questionable use of object and purpose. In the
end, their Honours also derived little assistance from it, instead drawing on this
merely as a means to support primarily textual interpretations.
B
Textual Interpretations of Art 2.5(a)
The primacy of the text of art 2.5(a) in their Honours’ interpretations of the
Australia-Hungary Extradition Treaty is clear. The joint judgment of Gummow,
Crennan, Kiefel and Bell JJ stated:
The ordinary, grammatical, meaningg of Art 2.5(a) is that extradition is not
to be granted unless the Minister is satisfied that the offence in relation
to which extradition is sought was an offence in the Requesting State at
the time of its alleged commission. If the pronoun ‘it’ is replaced with
the phrase for which it is surrogate, Art 2.5(a) reads ‘extradition may be
granted … irrespective of when the offence in relation to which extradition
is sought was committed, provided that the offence in relation to which
extradition is sought was an offence in the Requesting State at the time of
the acts or omissions constituting the offence’.76
Their Honours drew support from a comparison of art 2.5(a), which reads ‘was an
offence in the Requesting State’, with the text in art 2.5(b): ‘the acts or omissions
alleged would … have constituted an offence’.77 Their Honours did not find the
75
76
77
The judgment of McKerracher J makes repeated reference to the penalties being ‘different’: Zentai v
O’Connor [No 3] (2010) 187 FCR 495, 543 [200], 544–5 [210]–[211]. However, in the High Court this
was contradicted:
HEYDON J: What was the penalty for murder under the Hungarian Code?
MR LLOYD: … my instructions are that it was at the time death, but, neither of them are death
now. I think they are both imprisonment for life …
Transcript of Proceedings, Minister for Home Affairs v Zentai [2012] HCATrans 82 (28 March 2012).
Zentai (2012) 246 CLR 213, 241 [70] (emphasis added).
Ibid.
Zentai and The Troubles of Extradition
907
use of the phrase ‘an offence’ in both instances to be significant.78 Nor were their
Honours assisted by the provisions of art 2.2, including (a) ‘it shall not matter
whether the laws of the Contracting States place the acts or omissions constituting
the offence within the same category of offence or denominate the offence by the
same terminology’; and (b) ‘the totality of the acts or omissions alleged … shall
be taken into account’, which were found to be ‘pertinent to the application of
double criminality’ but of ‘no relevance’ to art 2.5(a).79 For Gummow, Crennan,
Kiefel and Bell JJ, the text and the purpose of art 2.5(a) prevented extradition to
face any charge which did not exist at the time of the conduct said to give rise to
it.80 Thus, s 22 of the Extradition Actt prevented Mr Zentai’s extradition to face
accusations under a retrospective war crimes law.
French CJ adopted a similar, albeit slightly more flexible, approach. As noted
above, his Honour agreed with the joint judgment that the purpose of art 2.5(a)
was the avoidance of extradition to face retrospective criminal charges; his
Honour similarly rejected the connection between arts 2.2 and 2.5.81 French
CJ also adopted an ‘ordinary grammatical construction’,82 but allowed greater
flexibility than the joint judgment:
the words ‘it was an offence in the Requesting State’ in Art 2.5(a) can be
construed broadly to encompass versions earlier in time of the ‘offence
in relation to which extradition is sought’ … That broad approach will
not encompass an offence created after the offending conduct which is
qualitatively different from the offence constituted by that conduct at the
time that that conduct was committed. … There is a point at which the
offence in the requesting state at the time of the relevant conduct cannot
be equated to the ‘offence in relation to which extradition is sought’.
Whatever that judgment may be in this case it cannot proceed on the basis
that if the conduct of the respondent constituted some species of criminal
offence at the time it was committed, that circumstance will be sufficient
to support a request for extradition in relation to any species of offence
later created by law and retroactively covering that conduct.83
On the facts, because the war crime was created in 1945 and the charged
conduct occurred in 1944, and the existent offences (including murder) were
all ‘qualitatively different’ from the 1945 war crime and not ‘versions earlier in
time’ of it, for French CJ art 2.5(a) of the extradition treaty prevented Mr Zentai’s
extradition to Hungary.
Heydon J’s textual interpretation was broader than that of the majority Justices:
‘All that matters under Art 2(5)(a) is that it was an offence — not necessarily
murder — in Hungary in 1944. [Even] [i]f the conduct was not murder, it was an
78
79
80
81
82
83
Australia-Hungary Extradition Treaty arts 2.5(a)–(b) (emphasis added).
Zentai (2012) 246 CLR 213, 241 [68].
Ibid 240–1 [68].
Ibid 227–8 [30]–[31].
Ibid 227 [30].
Ibid 228 [32]. This was a similar approach to that taken by Besanko J in the Full Court of the Federal
Court: O’Connor v Zentai (2011) 195 FCR 515, 530–1 [69]–[70].
908
Monash University Law Review (Vol 39, No 3)
offence’.84 Moreover, Heydon J regarded art 2.2 as relevant to the interpretation
of art 2.5.85 His Honour said:
the totality of the acts or omissions alleged must be taken into account …
[i]t does not matter that the constituent elements of a ‘war crime’ may be
greater in number than those of ‘murder’. It does not matter that they may
otherwise be different. The language of the Treaty directs attention to ‘the
totality of the acts or omissions alleged’.86
A key point of his Honour’s judgment was that the majority adopted an overly
technical interpretation of the text.87 His Honour remarked:
It is possible to draw attention to incongruities and infelicities in Art 2(5).
It is also possible to point to problems which would have been solved if
its terms were different. … An analyst could seek to draw conclusions
adverse to the appellants from that disconformity. That type of reasoning
is common in linguistic construction. It can be overdone. It is easy for
counsel to conduct a minute and leisurely examination of a document
years after it was drafted and ingeniously detect flaws in the drafting if it
is read one way. … This is a hypercritical approach. It is reminiscent of
the approach criminal defence counsel often take to a summing-up when
drafting a notice of appeal.
So far as that type of reasoning has merit, its merit certainly depends
on context. The present context concerns a bilateral treaty entered by
Australia and Hungary and negotiated by State representatives who, most
probably, did not share fluency in a common language. … As Deane J
said … ‘[i]nternational agreements are commonly “not expressed with the
precision of formal domestic documents as in English law”’.88
Heydon J therefore found that art 2.5(a) permitted Mr Zentai’s extradition to
Hungary to face a charge of war crimes, and would have upheld the Minister’s
appeal.89
There were three key differences between the Justices in Zentai. First, the majority
held that avoiding retrospective criminal liability was the purpose of art 2.5(a) of
the Australia-Hungary Extradition Treaty; Heydon J did not.90 Second, the joint
judgment adopted a stricter textual interpretation, concluding that ‘an offence’ in
art 2.5(a) meant ‘the offence in relation to which extradition is sought’,91 whereas
Heydon J interpreted ‘an offence’ to mean ‘any offence’.92 The middle ground
on this point was held by French CJ who found only ‘qualitatively different’
84
85
86
87
88
89
90
91
92
Zentai (2012) 246 CLR 213, 245 [81] (emphasis added).
Ibid 246–7 [86]–[88].
Ibid 247 [88].
Ibid 245 [82].
Ibid 247 [89]–[90], quoting Commonwealth v Tasmania (1983) 158 CLR 1, 261 (Deane J), in turn
quoting W Anstey Wynes, Legislative, Executive & Judicial Powers in Australia (Law Book, 5th ed,
1976) 299.
Zentai (2012) 246 CLR 213, 249 [99].
Ibid 228 [32] (French CJ), 240–1 [68] (Gummow, Crennan, Kiefel and Bell JJ), 243 [75] (Heydon J).
Ibid 241 [69]–[70].
Ibid 245 [81].
Zentai and The Troubles of Extradition
909
offences would be caught by art 2.5(a) and not ‘versions earlier in time’ of the
same offence.93 Third, only Heydon J regarded the broadening words of art 2.2 as
applicable to art 2.5.94
Of the differing textual interpretations, Heydon J’s objection to the strict
textual approach of the majority is persuasive, particularly when the broadening
provisions of art 2.2 are taken into account, and given that the prohibition of
retrospective criminal laws which influenced the majority’s approach was
arguably not applicable in Mr Zentai’s circumstances.
C
An Agreement as to the Meaning of Art 2.5(a)
The High Court in Zentai faced a curious argument that the meaning of art 2.5(a)
of the Australia-Hungary Extradition Treaty could be determined by the conduct
of the Hungarian and Australian governments in this case. The Minister argued
that Hungary’s request for the extradition of Mr Zentai, and the Minister’s s 22
determination approving extradition, supported the argument that art 2.5(a) did
not apply because these events were relevant to its interpretation as an instance of
subsequent agreement or practice coming within art 31(3) of the VCLT.
T 95
Notwithstanding the undoubted status and importance to the international law
of treaty interpretation of subsequent agreement and practice,96 in this case the
93
94
95
96
Ibid 228 [32].
Perhaps the most compelling interpretation of the extradition treaty was that adopted by North J,
dissenting in the Full Court of the Federal Court: see O’Connor v Zentai (2011) 195 FCR 515, 519–22
[16]–[30].
Zentai (2012) 246 CLR 213, 238 [64]. VCLT
T art 31(3) provides that, in the interpretation of a treaty:
There shall be taken into account, together with the context:
(a) any subsequent agreement between the parties regarding the interpretation of the
treaty or the application of its provisions;
(b) any subsequent practice in the application of the treaty which establishes the
agreement of the parties regarding its interpretation …
As Sir Gerald Fitzmaurice observed, ‘the way in which the treaty has actually been interpreted in practice
is evidence (sometimes the best evidence) of what its correct interpretation is’: Sir G G Fitzmaurice,
‘The Law and Procedure of the International Court of Justice: Treaty Interpretation and Certain Other
Treaty Points’ (1951) 28 British Year Book of International Law 1, 9. Similarly Aust observed that
subsequent practice is ‘a most important element in the interpretation of any treaty’: Anthony Aust,
Modern Treaty Law and Practice (Cambridge University Press, 2ndd ed, 2007) 241. Dörr said it is ‘of
utmost importance’: Oliver Dörr, ‘Article 31: General Rule of Interpretation’ in Oliver Dörr and Kirsten
Schmalenbach (eds), Vienna Convention on the Law of Treaties: A Commentary (Springer, 2012) 521,
554. Gardiner writes that ‘treaties embody the common understanding of the parties to them. Hence
concordant practice of the parties is best evidence of their correct interpretation’: Richard K Gardiner,
Treaty Interpretation (Oxford University Press, 2008) 225. See also International Status of South-West
Africa ((Advisory Opinion) [1950] ICJ Rep 128, 135–6; Legal Consequences for States of the Continued
Presence of South Africa in Namibia (South West Africa) notwithstanding Security Council Resolution
276 (1970) ((Advisory Opinion) [1971] ICJ Rep 16, 22; Hall, above n 41, 109–10 [2.79]; David Harris,
Cases and Materials on International Law (Sweet & Maxwell, 7th ed, 2010) 678–9; Ian Brownlie,
Principles of Public International Law (Oxford University Press, 7th ed, 2008) 633–4; Sir Ian Sinclair,
The Vienna Convention on the Law of Treaties (Manchester University Press, 2ndd ed, 1984) 135–8;
Hazel Fox, ‘Article 31(3)(a) and (b) of the Vienna Convention and the Kasikili/Sedudu Island Case’ in
Malgosia Fitzmaurice, Olufemi Elias and Panos Merkouris (eds), Treaty Interpretation and the Vienna
Convention on the Law of Treaties: 30 Years On (Martinus Nijhoff Publishers, 2010) 59, 60–4.
910
Monash University Law Review (Vol 39, No 3)
Minister’s argument was specious,97 as their Honours realised. The Chief Justice
wrote:
It may be debatable whether the making of a request for extradition and
the accession to that request on the basis of a common opinion about the
interpretation of the Treaty requires that opinion to be taken into account
in interpreting the Treaty.98
As Crennan J had perceptively asked at the hearing:
Just on the practice point, would not practice normally suggest, well, first,
consistency in relation to a number of occasions so that both number —
repeat occasions and consistency will give rise to a practice? Can you
tease a practice out of one instance?99
Their Honours were correct to reject the Minister’s argument. Under international
law, the mere conduct of Hungary in requesting Mr Zentai’s extradition, and of
Australia in agreeing to it, is not capable of giving rise to ‘subsequent practice’
within art 31(3). Not only is it clear that ‘“practice” cannot be established by one
isolated incident’,100 but ‘[t]he essence of it is what can be shown to have been
done systematically or repeatedlyy in implementation and application of a treaty’.101
Moreover, subsequent practice is not made out by a subjective statement of position
in one particular instance, but is relevant only as ‘objective evidence of [the parties’]
understanding as to the meaning’ of a provision established over time.102
In Zentai, there was no relevant subsequent practice in the interpretation of
the Australia-Hungary Extradition Treaty to assist in the interpretation of art
2.5(a). The supposed agreement between Hungary and Australia in this case
was irrelevant to the interpretation of the extradition treaty as a matter of the
international law of treaty interpretation.
D A Presumption in Favour of Liberty?
An alternative perspective on the Court’s approach in Zentai may be gained from
a comment made by French CJ at the hearing:
97
Written submissions filed on behalf of Mr Zentai made this point emphatically: ‘In effect, therefore, the
Minister seeks to establish the validity of his action by reliance on the action itself; an exercise in both
self-levitation and self-empowerment’: Charles Zentai, ‘First Respondent’s Submissions’, Submission
in Minister for Home Affairs (Cth) v Zentai, P56 of 2011, 10 February 2012, 12 [34].
98 Zentai (2012) 246 CLR 213, 229 [36].
99 Transcript of Proceedings, Minister for Home Affairs (Cth) v Zentai [2012] HCATrans 82 (28 March
2012).
100 Dörr, above n 96, 556. See also Sinclair, above n 96, 137, where it is said that ‘practice … cannot in
general be established by one isolated fact or act or even by several individual applications’.
101 Gardiner, above n 96, 226 (emphasis added). Similarly, it must be ‘constant and repeated’: at 231. Dörr
said that it must be ‘sufficient to establish a discernable pattern of behaviour’ and ‘must constitute a
sequence of acts or pronouncements’: Dörr, above n 96, 556. Similarly, Sinclair said that ‘[t]he value
and significance of subsequent practice will naturally depend on the extent to which it is concordant,
common and consistent. A practice is a sequence of facts or acts’: Sinclair, above n 96, 137.
102 Dörr, above n 96, 554.
Zentai and The Troubles of Extradition
911
the problem is … you have a treaty which is not expressed with precision,
as many treaties are not, and that treaty contains in one of its provisions
a construction or choice between two constructions, one of which would
result in somebody being arrested, locked up and then sent overseas to be
tried, and another of which would not.103
One additional explanation of the majority position might be that, in case of
ambiguity, their Honours adopted the interpretation in favour of individual
liberty.104
The case, however, was not one to which the presumption that penal provisions
are strictly construed105 should be applied. As the House of Lords has held,
the common law presumptions will often be inapposite in the construction of
extradition treaties, which are
intended to serve the purpose of bringing to justice those who are guilty
of grave crimes committed in either of the contracting states. To apply
to extradition treaties the strict canons appropriate to the construction of
domestic legislation would often tend to defeat rather than to serve this
purpose.106
Indeed, counsel for Mr Zentai acknowledged the inappropriateness of applying
presumptions of statutory interpretation: ‘the people who are negotiating the
treaty and drafting it may not have a copy of Pearce and Geddes at their elbow
and cannot be taken to have been proceeding according to the common law on
presumptions of construction’.107
Further, as Gleeson CJ explained in Vasiljkovic:
Plainly, extradition has serious implications for the human rights … of the
person who is the subject of a request for surrender. … The interference
with personal liberty involved in detention during the extradition process
… and in involuntary delivery to another country and its justice system is
not undertaken as a form of punishment. No doubt, to the person involved,
some of its practical consequences may be no different from punishment, but
the purpose is not punitive. To repeat, the process involves no adjudication
of guilt or innocence. It is undertaken for the purpose of enabling such
103 Transcript of Proceedings, Minister for Home Affairs (Cth) v Zentai [2012] HCATrans 82 (28 March
2012).
104 See, eg, Pearce and Geddes, above n 41, 297–9 [9.9], 309–10 [9.32].
105 The classic statement of that presumption remains the formulation of Brett J in Dickenson v Fletcher
(1873) LR 9 CP 1, 7:
Those who contend that the penalty may be inflicted, must shew that the words of the Act
distinctly enact that it shall be incurred under the present circumstances. They must fail, if the
words are merely equally capable of a construction that would, and one that would not, inflict
the penalty.
106 Belgium v Postlethwaite [1988] AC 924, 947 (Lord Bridge), quoted with approval in O’Connor v Zentai
(2011) 195 FCR 515, 519 [12] (North J).
107 Transcript of Proceedings, Minister for Home Affairs (Cth) v Zentai [2012] HCATrans 82 (28 March
2012) (G R Kennett SC). McKerracher J in the Federal Court had, however, relied upon the presumption
against retrospective operation: Zentai v O’Connor [No 3] (2010) 187 FCR 495, 541 [189], 542 [193].
912
Monash University Law Review (Vol 39, No 3)
an adjudication to be made in a foreign place, according to foreign law,
in circumstances where Australia has no intention itself of bringing the
person to trial for the conduct of which the person is accused.108
Extradition laws do not themselves serve a purpose of punishment, but exist
instead ‘to restore fugitive criminals to the jurisdiction of a court competent
by municipal and international law to try them’.109 It is not clear whether any
presumption in favour of individual liberty actually informed their Honours’
interpretations in Zentai; it is, however, likely that the presumption itself is not
applicable in the context of extradition.110
E
Conclusion: Art 2.5(a) Prevents the Extradition of
Mr Zentai
The majority of the High Court concluded that Mr Zentai could not be extradited to
Hungary because the war crimes offence for which he was wanted for questioning
came into existence in 1945 and was thus not ‘an offence in the Requesting State
at the time of the acts or omissions constituting the offence’, notwithstanding
that the alleged conduct would have been undoubtedly criminal as murder. This
interpretation was based on a reading of the text and upon an understanding of
the purpose of art 2.5(a).
The Court’s reliance on a strict textual interpretation, its failure to consider the
relevance of the purpose of extradition arrangements as a whole, and its use of the
arguably inapplicable international law prohibition of retrospective criminal law
in the interpretation of art 2.5(a) are all matters of concern regarding the approach
taken. Their Honours were, however, correct to reject the Commonwealth’s
argument that there was an agreement that art 2.5(a) did not apply in this case,
which was not a legitimate invocation of the use of subsequent agreement and
practice in the international law of treaty interpretation.
V CONCLUSION
The High Court’s dismissal of the Minister’s appeal in Zentai meant that Mr Zentai
would not be extradited to Hungary to face war crimes charges. The decision
in Zentai is a legal milestone: in all likelihood it brings to an end Australia’s
contributions to bringing to justice alleged war criminals from the Second World
108 Vasiljkovic (2006) 227 CLR 614, 629–30 [34] (Gleeson CJ).
109 Shearer, Extradition in International Law, above n 26, 67. See also Aughterson, above n 4, 33.
110 The Court in Oates v A-G (Cth) (2003) 214 CLR 496 was asked to apply the presumption in the context
of extradition, but found it inapplicable on the facts: at 513 [45] (Gleeson CJ, McHugh, Gummow,
Kirby, Hayne and Heydon JJ). The Court did not indicate whether it could apply to extradition more
generally.
Zentai and The Troubles of Extradition
913
War — leaving a controversial record.111 Curiously, the Court’s decision has the
practical effect that Australia cannot extradite Mr Zentai to face a charge of war
crimes created in Hungary in 1945, although Australia could itself charge Mr
Zentai (over the same conduct) with a war crimes offence created in 1988,112 the
validity of which was upheld by the High Court in Polyukhovich.113
Zentai has a number of important consequences. First, the extraordinary time
and resources devoted to the extradition process raise, once again, the issue
of whether Australia’s extradition arrangements require reforms to promote
efficiency. At present, the human rights protections contained in extradition
treaties fall to be considered by the Minister under s 22 of the Extradition Act. It
may be appropriate to make these matters part of the s 19 judicial determination
so that they are resolved earlier in the extradition process.
Second, the strict textual interpretation that ‘an offence’ in art 2.5(a) means the
specific offence for which extradition is sought has implications for all similarlydrafted extradition treaties. Under these treaties, no person will now be extradited
to face trial for a retrospective criminal offence, even if the offence is only
technically different from existing offences. In light of Zentai, the Commonwealth
may need to consider amendments to the relevant treaties.
There is considerable resonance of Deane J’s caution in Riley v Commonwealth
that the double criminality rule in extradition law should not be applied to the
detriment of the overall purpose of the extradition regime, lest its utility in
protecting the rights of the accused
be outweighed by the impediment which it represents to the advancement
of criminal justice if its content is defined in over-technical terms which
would preclude extradition by reason of technical differences between
legal systems, notwithstanding that the acts alleged against the accused
involve serious criminality under the law of both requesting and requested
states.114
Mr Zentai was accused of unlawful killing, albeit as a war crime rather than
murder. It was nonetheless one of the most serious crimes in both Australia and
111 On this point, Efraim Zuroff, Israeli director of the Simon Weisenthal Centre claimed: ‘In practical
terms, it signals a dismal conclusion to Australia’s totally unsuccessful efforts to bring to justice any
of the numerous Nazi war criminals who found refuge in the country’: Simon Weisenthal Centre,
‘Wiesenthal Center: Decision to Block Extradition of Zentai to Hungary for War Crimes Seals Australia’s
Total Failure to Bring Nazi War Criminals to Justice’ (Press Release, 15 August 2012) <http://www.
wiesenthal.com/site/apps/ nlnet/content2.aspx?c=lsKWLbPJLnF&b=4441467&ct=12110281>.
112 War Crimes Act 1945 (Cth) s 9, amended by War Crimes Amendment Act 1988 (Cth) s 5. At first instance
in the Federal Court, McKerracher J had accepted Mr Zentai’s argument that art 2.5(a) was inconsistent
with Australia’s attitude against retrospective criminal law, and referred to the non-retrospective war
crimes offences enacted prior to the entry into force of the Rome Statute: Zentai v O’Connor [No 3]
(2010) 187 FCR 495, 543 [202]. Strangely, Polyukhovich v Commonwealth (1991) 172 CLR 501
(‘Polyukhovich’) was not discussed in this context.
113 (1991) 172 CLR 501. There had, however, been an assessment by the Commonwealth Director of Public
Prosecutions, who had declined to prosecute Mr Zentai in Australia because ‘without any testimony
from living witnesses, a prima facie case did not exist under the War Crimes Act 1945 (Cth)’: Tully,
above n 20, 269.
114 (1985) 159 CLR 1, 17.
914
Monash University Law Review (Vol 39, No 3)
Hungary at all relevant times. The majority’s interpretation of art 2.5(a) of the
Australia-Hungary Extradition Treaty undoubtedly protected Mr Zentai from
what he feared would be an unfair trial in Hungary.115 However, it achieved
this result based on an overly technical interpretation of the extradition treaty
which failed to take into account the important role of extradition in facilitating
international cooperation in the apprehension and surrender for trial of those
accused of serious criminal offences.
115 See, eg, Tully, above n 20, 270; Zentai v O’Connor [No 3] (2010) 187 FCR 495, 551–72 [260]–[321].